More crossover help please

Y

yzracer14

New Member
For the past few weeks I’ve been researching crossovers and everything related. But I’m finding that as I go, the more advanced it gets and I’m now at the point where I need some help because I'm not understanding.

I’ve tried every calculator that I’ve found through Google and I cannot replicate the values in the original crossover network. Not even close. Then again, not knowing the current inductor values doesn’t really help me either. To me, I can’t justify dropping $100+ on high quality capacitors without knowing that as much of the frequency as possible is being played by one or more speakers. Why does that worry me? Because with the current values right now, there seem to be a few frequencies that I can’t hear at all through my CW S-1’s but I can hear perfectly through my TV’s speakers. it really bugs me. Not to mention that the previous owner put two different 12" woofers in it so the sound is pretty off.

So what I would like to do is recalculate the crossover network and replace every component including the inductors to make 100% sure that every possible frequency is covered. I know that impedance varies while the speaker is in use but when calculating the new crossover values, should I use the ohm values written on the back of the speaker, or the values that I get when read with a DMM? The reason I ask is because the tweeter reads 4 ohms on the back of the speaker, but when read with a DMM, it reads 2.2 ohms. Unfortunately I haven’t been able to locate any data specs or frequency response curves for the mid or tweeter and yes I already contacted Cerwin Vega directly. The only spec sheets I have are for the 12” woofer because I’m replacing them with a new pair.

I have plenty of other questions but I’ll hold off on those for now because my questions may change depending on the answers I get, assuming I get any. I thank you guys in advance. So far I’ve had nothing but great help and it’s greatly appreciated.

EDIT: Added original crossover schematic for reference if needed. Much better than my other attempts I think.
 

Attachments

  • Stock XO.jpg
    Stock XO.jpg
    48.7 KB · Views: 26
Register to hide this ad
Replace the caps with the stock values and be done with it; it's unlikely there's anything to "upgrade" about the inductors or resistors.

You're talking about redesigning the speakers here, and that's not going to happen successfully without the requisite knowledge, skill, and measurement tools.

The calculators are based upon textbook models, and you'd be hard-pressed to find a manufacturer using those values in a product, because there are no drivers or combinations of drivers that perform according to theory acoustically.

For starters, the stated impedance is "nominal," and what you're measuring with a DMM is DC resistance at zero frequency, not the same. For the calcs to be right, you must use the measured or modeled impedance at the design filter frequency.

Bottom line, it a hobby, and fun kerfutzing with this stuff and learning how it works, but don't think for a minute that you're going to produce an optimum result without investing the requisite effort in that accomplishment. You're well along into this, apparently, and I certainly encourage you to carry on with the pursuit; just be aware that the achievable objective may be considerably different from that which you presently envisage.... :yes:
 
Often times I need to brought back down to earth because I tend to like to over-do things. You’ve done that and more; in a good way of course.

I didn’t intend to redesign the speakers as I have the originally designed crossover points and I planned to stick with those. But without knowing the original inductor values or which type of 2nd order it uses, it would pretty much force me to re-design the entire crossover ( i see that now ) and I don’t think I’m up for that task due to my lack of experience and knowledge at this point in time. So for now I’ll just replace the stock values with higher quality components as suggested.

However, I will have to change the zobel because I’m replacing the 12” speakers and they have different specs than the originals. The 40uF cap going straight to ground will remain the same value so it shouldn't affect the crossover points but when plugging the speaker specs into the zobel calculators it gives me a 5 ohm resistor value and an 80uF capacitor value. So those components will get changed.

Like I said in my first post, I do have more questions, although a lot were eliminated.

1) With the 12” speaker circuit; will upgrading the cap that goes straight to ground to a film cap give me any sonic benefits at all? What about with the zobel?

2) If I run out of room in the originally allocated crossover space; is it acceptable to wire the 12” woofer components up at the speaker connections or is it best to keep them in the original space close to the other components? Sonically, are there any positives and/or negatives to either configuration?

3) I plan on doing some experimenting with different types of caps as bypass capacitors. I know it’s said that high quality components don’t need them but they’re cheap enough to experiment with and it won’t hurt anything. But I wanted to get some opinions on some of the things I plan on doing. FYI – I don’t plan on doing ANY experimenting until I get the crossover fully recapped and working. One step at a time.

The main cap for my experimentation will be a film cap and I plan on trying the following as bypass caps: other types of film, elec., ceramic, and a straight wire. There might be more but those are the ones that pop into my head at the moment. I’ve done a lot of cap experimentation in guitar effects pedals with good results so I’m looking forward to this.

My theory with the straight wire is that all caps give some sort of coloration to the sound. So, why not by-pass with a straight wire to get some more ‘non-colored’ sound through to your ears? Take my crossover for example; with the tweeter I have a 3.5uF cap. If I raise that to a 7uF cap and bypass it with a bare wire that would bring it back down to 3.5uF right? Or am I wrong in my line of thinking here?

Now that I read it typed up, my theoory with the bare wire doesn't sound right but I'll leave it anyways to get some other thoughts on it.
 
1) Plenty of folks say upgrading caps in the lowpass is not necessary. I replace NPEs with metalized films, if the sizes are reasonable, but it's just Daytons or Solens, nothing allegedly "better."

2) Proximity to other crossover components is not an issue.

3) Zobels are often employed as response shaping filters, not merely impedance compensation. The calcs may not give the "right" values for any particular driver and lowpass combination.

4) I follow current JBL practice with bypass caps, 0.01 uF film & foil, Thetas if I'm feeling flush. I don't bypass Solens.

5) Your straight wire thing makes no sense as I read it. If you really want to play here, do the charge-coupled bias thing.... :yes:
 
yzracer14 said:
Take my crossover for example; with the tweeter I have a 3.5uF cap. If I raise that to a 7uF cap and bypass it with a bare wire that would bring it back down to 3.5uF right? Or am I wrong in my line of thinking here?

Now that I read it typed up, my theoory with the bare wire doesn't sound right but I'll leave it anyways to get some other thoughts on it.
Click to expand...

Like Zilch said, your bare wire by-pass makes no sense whatsoever.

What it does is short the cap resulting in NO high pass filter at all. That means your tweeter will get a full range signal and it start producing the lower frequencies in conflict with the other drivers.

If your speakers currently are hifi, then you will turn them into midfi or worse.

As for coloration in caps? The primary function of capacitors in an xover is frequency selection (in some xovers they also do some wave shaping). That is, we limit the frequencies to drivers so that they operate in the linear range over which they were designed.

Regards,
Jerry
 
I saved the website to my favorites and read through a lot of things. I focused heavily on the zobel network and I understand it and calculated the capacitor for my intended new 12” speaker ( 1.7 inductance and 4.8 ohm minimum capacitance) to be 74uF.

However, there are a few things that I don’t understand that aren’t covered in that article, or other articles I found on his site for that matter. The first one being the value of the resistor; how do I determine the value of the resistor to use? The next issue is his comment that, “the zobel network must be treated as part of the loudspeaker. All measurements or calculations for the crossover network must include the zobel network and loudspeaker driver combined.” From what I understand, it means that the capacitor value for the zobel plays a part in the lowpass section of the crossover. So in my specific crossover where the combined value of the caps for the 12” speaker lowpass is 80uF, in order to keep the crossover frequency the same I would have to reduce the other capacitor to 6uF to bring the total to 80uF. Again, that’s what I understood it as but I wouldn’t be here if I wasn’t sure. Is my understanding correct or is there something I’m missing?

Also, is there anybody on this forum that knows what the original 12” speaker’s nominal ohm value was? As I’ve said before, one of the original owners replaced both 12” speakers and I have no idea if they are correct. I do know that the model # of the original speaker was 122S4. I googled it and came up with nothing as far as specs go. I’ve also searched AK and messaged other AK’ers who’ve owned S-1’s with no luck as well as contacting CV themselves with no usable answer.
 
The Zobel is altering the impedance that the rest of the crossover "sees". Some impedance compensation circuits flatten out the rising impedance that all drivers have and some knock down the impedance at the drivers resonance. If you used both, you could somewhat approximate a resistive load when modeling the crossover (I say somewhat because it will still remain reactive, it will just have a flatter impedance curve). So, if you had a woofer that started to have significantly rising impedance near the crossover point, let's say it's at 32 ohms and rising sharply at 3000hz, the crossover values needed would be signfiicantly different than if you used a zobel to keep the impedance at 8 ohm at 3000hz. They are just making sure that you use the altered impedance curve rather than the original drivers impedance curve when modeling/designing.
 
dnewma04 said:
Crap. I hate writing longs posts when simple ones are more effective.
Click to expand...
I added "Model" after reading your post. ;)

Don't know if it's apparent to everyone, but there's an increased interest in objectification of speakers here on AK of late, and figuring them out.

[Ray told everyone that we had cookies.... :lmao: ]
 
I've noticed it, too. It'll keep getting easier as people are more effectively able to understand how the measurements correlate to what they hear. Measurements aren't as easily dismissible with more understanding.
 
You have gone too far. I finally decided on a preamp (the Tascom 144), ordered it and went to PE to order the Behringer mic. OUT OF STOCK! Likewise ditto two other stores that normally carry it. Too many people trying to analyze the noise a speaker makes when it hertz!

Or something like that... The cookies are good though. :tongue:
 
What we do on AK is sometimes a challenge to PE's inventory management team. ;)

I hope everyone's using the PE link at the bottom of these pages so AK gets its full measure of due blame for this.... :thmbsp:
 
yzracer14 said:
My theory with the straight wire is that all caps give some sort of coloration to the sound. So, why not by-pass with a straight wire to get some more ‘non-colored’ sound through to your ears? Take my crossover for example; with the tweeter I have a 3.5uF cap. If I raise that to a 7uF cap and bypass it with a bare wire that would bring it back down to 3.5uF right? Or am I wrong in my line of thinking here?

Now that I read it typed up, my theoory with the bare wire doesn't sound right but I'll leave it anyways to get some other thoughts on it.
Click to expand...

At audio frequencies that bare wire will simply short out the cap. At frequencies above 30mhz the capacitance of the wire starts making a difference and has to be compensated for. Stick with those 0.01 mf caps for bypassing.
 
You must log in or register to reply here.
Back
Top Bottom